• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Ivybridge Overclocking and OCCT

KingDazza

KingDazza

KingDazza

Associate
Joined
30 Dec 2013
Posts
542
Location
England
Hi,

To avoid upgrading to Kabylake. I decided after 4 years I'd overclock my i7 3770k. Currently running at a permanent 4.2Ghz @ fixed vcore 1.17.

I left it running OCCT CPU Large overnight and it hit an error after just under 5 hours. No BSOD or gaming issues. So should I just increase the vcore to say 1.19v and try again tonight? How long is stable? Is OCCT a reliable and accurate way of testing.

Any thoughts and advice most welcome. :)
 
A lot of people will say that the best stability testing is to use the pc for the tasks it is intended for. Asus realbench is a good tester of stability for normal uses. A voltage of only 1.17v seems rather low for 4.2Ghz so it may not be enough. The thing is that if I were you I would be thinking I have some sort of golden chip there and would be trying for a lot more than 4.2Ghz.
 
pastymuncher said:
A lot of people will say that the best stability testing is to use the pc for the tasks it is intended for. Asus realbench is a good tester of stability for normal uses. A voltage of only 1.17v seems rather low for 4.2Ghz so it may not be enough. The thing is that if I were you I would be thinking I have some sort of golden chip there and would be trying for a lot more than 4.2Ghz.
Click to expand...

Thanks. I have real bench and it ran on that for several hours fine.

It depends where you read though. Stock volts are 1.15 to 1.16 and many people reckoned they hit 4.2 at stock volts... I would say its approaching good stability now and only needs a bit more tweaking. So perhaps a better question would be if adjusting by 0.02 volts is a good way to go and better than 0.005 at a time with tests taking how ever many hours we deem appropriate.
 
You should be able to up the voltage just a little and get up to 4.5. That would probably worth trying. IMO.
 
djbully said:
You should be able to up the voltage just a little and get up to 4.5. That would probably worth trying. IMO.
Click to expand...

Yeah, unfortunately I'm limited by temps. This chip runs hot regardless of cooler, so would likely need to delid which I'd like to avoid at this time. Just trying to get 4.2 fully stable at this point. :)
 
andy_mk3 said:
I don't worry about stress tests any more. If it can handle a 6 hour gaming session then it's stable to me.
Click to expand...

Thats a philosophy I've been hearing more and more. It ran Attila Total War for 7 hours with no issues yesterday. Still it would have been nice if OCCT hadn't thrown a wobbly after 5 hours.
 
andy_mk3 said:
I don't worry about stress tests any more. If it can handle a 6 hour gaming session then it's stable to me.
Click to expand...

Depends on what you require from the PC - its quite possible to be game "stable" but have corruption creep into big file copies, etc. etc.

I usually give it 2-3 hours or so of XTU and again with an encoding type stress test - so far not seen anything that passes that and subsequently fails OCCT or the like.
 
I'm running my 3770k here, delidded with CLU 1.35v under custom water.

I'm not sure what cooler you're using but if it is half decent up to 1.3v should be ok :)

I definitely would be careful with it but I always found IBT to be the best thermal and stability test. Some are dead against it, but it works well for me.
 
Bluntwrapped said:
I'm running my 3770k here, delidded with CLU 1.35v under custom water.

I'm not sure what cooler you're using but if it is half decent up to 1.3v should be ok :)

I definitely would be careful with it but I always found IBT to be the best thermal and stability test. Some are dead against it, but it works well for me.
Click to expand...

Guess I'm just unlucky but with my chip I got the same temps using 3 different air coolers with the best being a top tier Noctua - identical temps. It really needs delidding. But for now I'd be happy just getting 4.2 fully stable. Passed IBT fine on maximum. Its just the overnight OCCT that failed.
 
I've been playing with the 3770k in my son's pc. Using Realbench encoding. It throws up errors quicker than most other programs and you can check in the windows event logs for WHEA errors to get the voltage dialled in.
 
Last edited:
Rroff said:
Depends on what you require from the PC - its quite possible to be game "stable" but have corruption creep into big file copies, etc. etc.
Click to expand...

This. I would only use gaming as a stability test if I could afford to (in principal) lose all data and/or reinstall the OS periodically.

Too much chance of corruption - not a high chance, but not zero as you get with a stock CPU either.
 
I just run mine at 4.3 with max 1.184v I think it is. Will do 4.5 at about 1.26 but prefer the lower temps and noise due to cooler. Need a bigger case :o

Used a combination of realbench, prime, benchmarks and games.
 
Last edited:
Drollic said:
I just run mine at 4.3 with max 1.184v I think it is. Will do 4.5 at about 1.26 but prefer the lower temps and noise due to cooler. Need a bigger case :o

Used a combination of realbench, prime, benchmarks and games.
Click to expand...

Thanks, yep similar situation. I've upped to 1.19v for now and will run OCCT over night again. Going to 4.4 or above just requires too much volts and heat.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom